Daily Papers of 2026-05-20

  1. Anti-Self-Distillation for Reasoning RL via Pointwise Mutual Information 191 upvotes, #1 of 2026-05-20
  2. AutoResearchClaw: Self-Reinforcing Autonomous Research with Human-AI Collaboration 182 upvotes, #2 of 2026-05-20
  3. When Vision Speaks for Sound 148 upvotes, #3 of 2026-05-20
  4. Active Learners as Efficient PRP Rerankers 96 upvotes, #4 of 2026-05-20
  5. OpenComputer: Verifiable Software Worlds for Computer-Use Agents 57 upvotes, #5 of 2026-05-20
  6. GoLongRL: Capability-Oriented Long Context Reinforcement Learning with Multitask Alignment 56 upvotes, #6 of 2026-05-20
  7. Process Rewards with Learned Reliability 52 upvotes, #7 of 2026-05-20
  8. EnvFactory: Scaling Tool-Use Agents via Executable Environments Synthesis and Robust RL 48 upvotes, #8 of 2026-05-20
  9. Harnessing LLM Agents with Skill Programs 34 upvotes, #9 of 2026-05-20
  10. CogOmniControl: Reasoning-Driven Controllable Video Generation via Creative Intent Cognition 34 upvotes, #9 of 2026-05-20
  11. Aurora: Unified Video Editing with a Tool-Using Agent 29 upvotes, #11 of 2026-05-20
  12. Artifact-Bench: Evaluating MLLMs on Detecting and Assessing the Artifacts of AI-Generated Videos 22 upvotes, #12 of 2026-05-20
  13. OmniGUI: Benchmarking GUI Agents in Omni-Modal Smartphone Environments 16 upvotes, #13 of 2026-05-20
  14. ThoughtTrace: Understanding User Thoughts in Real-World LLM Interactions 16 upvotes, #13 of 2026-05-20
  15. Interactive Evaluation Requires a Design Science 14 upvotes, #15 of 2026-05-20
  16. CEPO: RLVR Self-Distillation using Contrastive Evidence Policy Optimization 14 upvotes, #15 of 2026-05-20
  17. MSAVBench: Towards Comprehensive and Reliable Evaluation of Multi-Shot Audio-Video Generation 14 upvotes, #15 of 2026-05-20
  18. SENSE: Satellite-based ENergy Synthesis for Sustainable Environment 13 upvotes, #18 of 2026-05-20
  19. Video Models Can Reason with Verifiable Rewards 11 upvotes, #19 of 2026-05-20
  20. PixVerve: Advancing Native UHR Image Generation to 100MP with a Large-Scale High-Quality Dataset 11 upvotes, #19 of 2026-05-20
  21. Semantic Generative Tuning for Unified Multimodal Models 10 upvotes, #21 of 2026-05-20
  22. Fast 4D Mesh Generation by Spatio-Temporal Attention Chains 10 upvotes, #21 of 2026-05-20
  23. RT-Splatting: Joint Reflection-Transmission Modeling with Gaussian Splatting 9 upvotes, #23 of 2026-05-20
  24. Overcoming Catastrophic Forgetting in Visual Continual Learning with Reinforcement Fine-Tuning 8 upvotes, #24 of 2026-05-20
  25. Delta Attention Residuals 8 upvotes, #24 of 2026-05-20
  26. Code-Guided Reasoning for Small Language Models: Evaluating Executable MCQA Scaffolds 7 upvotes, #26 of 2026-05-20
  27. PEEK: Context Map as an Orientation Cache for Long-Context LLM Agents 7 upvotes, #26 of 2026-05-20
  28. Draft Less, Retrieve More: Hybrid Tree Construction for Speculative Decoding 7 upvotes, #26 of 2026-05-20
  29. TideGS: Scalable Training of Over One Billion 3D Gaussian Splatting Primitives via Out-of-Core Optimization 7 upvotes, #26 of 2026-05-20
  30. Zero-Shot Sim-to-Real Robot Learning: A Dexterous Manipulation Study on Reactive Catching 6 upvotes, #30 of 2026-05-20
  31. Context Memorization for Efficient Long Context Generation 6 upvotes, #30 of 2026-05-20
  32. Matérn Noise for Triangulation-Agnostic Flow Matching on Meshes 6 upvotes, #30 of 2026-05-20
  33. optimize_anything: A Universal API for Optimizing any Text Parameter 6 upvotes, #30 of 2026-05-20
  34. Not Every Rubric Teaches Equally: Policy-Aware Rubric Rewards for RLVR 6 upvotes, #30 of 2026-05-20
  35. Mid-Training with Self-Generated Data Improves Reinforcement Learning in Language Models 5 upvotes, #35 of 2026-05-20
  36. Echo-Forcing: A Scene Memory Framework for Interactive Long Video Generation 5 upvotes, #35 of 2026-05-20
  37. ESI-Bench: Towards Embodied Spatial Intelligence that Closes the Perception-Action Loop 5 upvotes, #35 of 2026-05-20
  38. Where Does Authorship Signal Emerge in Encoder-Based Language Models? 5 upvotes, #35 of 2026-05-20
  39. Stage-adaptive Token Selection for Efficient Omni-modal LLMs 5 upvotes, #35 of 2026-05-20
  40. DocAtlas: Multilingual Document Understanding Across 80+ Languages 4 upvotes, #40 of 2026-05-20
  41. Editor's Choice: Evaluating Abstract Intent in Image Editing through Atomic Entity Analysis 4 upvotes, #40 of 2026-05-20
  42. Why Do Reasoning Models Lose Coverage? The Role of Data and Forks in the Road 4 upvotes, #40 of 2026-05-20
  43. Omni-DuplexEval: Evaluating Real-time Duplex Omni-modal Interaction 4 upvotes, #40 of 2026-05-20
  44. Ethical Hyper-Velocity (EHV): A Provably Deterministic Governance-Aware JIT Compiler Architecture for Agentic Systems 4 upvotes, #40 of 2026-05-20
  45. Language-Switching Triggers Take a Latent Detour Through Language Models 4 upvotes, #40 of 2026-05-20
  46. CopT: Contrastive On-Policy Thinking with Continuous Spaces for General and Agentic Reasoning 4 upvotes, #40 of 2026-05-20
  47. Bug or Feature^2: Weight Drift, Activation Sparsity, and Spikes 3 upvotes, #47 of 2026-05-20
  48. Be Kind, Rewrite: Benign Projections via Rewriting Defend Against LLM Data Poisoning Attacks 3 upvotes, #47 of 2026-05-20
  49. Computer Science Conferences Should Require Nonrepudiable Experimental Results 2 upvotes, #49 of 2026-05-20
  50. Base Models Look Human To AI Detectors 2 upvotes, #49 of 2026-05-20
  51. S-Bus: Automatic Read-Set Reconstruction for Multi-Agent LLM State Coordination 1 upvotes, #51 of 2026-05-20
  52. SAGA: A Sequence-Adaptive Generative Architecture for Multi-Horizon Probabilistic Forecasting with Adaptive Temporal Conformal Prediction 1 upvotes, #51 of 2026-05-20
  53. RoPE Distinguishes Neither Positions Nor Tokens in Long Contexts, Provably 0 upvotes, #53 of 2026-05-20
  54. SceneCode: Executable World Programs for Editable Indoor Scenes with Articulated Objects 10 upvotes, #53 of 2026-05-20

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.